Automatic Descriptive Exam Answer Grading Using Text Information Analysis and Retrieval - دانشکده فنی و مهندسی
Automatic Descriptive Exam Answer Grading Using Text Information Analysis and Retrieval
نوع: Type: thesis
مقطع: Segment: masters
عنوان: Title: Automatic Descriptive Exam Answer Grading Using Text Information Analysis and Retrieval
ارائه دهنده: Provider: Amin Sharifi nia
اساتید راهنما: Supervisors: Dr. Muharram Mansoorizadeh
اساتید مشاور: Advisory Professors: Dr. Farhad Seraji
اساتید ممتحن یا داور: Examining professors or referees: Dr. Mir Hossein Dezfoulian Dr. Hassan Khotanlou
زمان و تاریخ ارائه: Time and date of presentation: April 21, 2020
مکان ارائه: Place of presentation: http://vc.basu.ac.ir/eng-thesis06
چکیده: Abstract: Testing is a measurement tool in education and psychology. In measurement, we determine the properties or attributes of objects and people and report them as numbers. Different instruments are used to measure the different characteristics of objects and people. For example, meters and scales are used to measure the length and weight characteristics of objects. In education and psychology, tests are used to measure the psychological and behavioral characteristics of individuals. The test consists of a set of questions that are given to the person to answer. One of the most important aspects of learning is the process of assessing learners' knowledge While some forms of assessment are computerized and do not require complex text comprehension (e.g., multiple choice questions or true / false questions) that can be easily graded by the system. But in some cases, students' responses consist of free text that needs to be analyzed. Therefore, in this study, we created a system to correct the answers to descriptive questions. We used text retrieval methods in our system. And after the necessary preprocessing on the student responses by the clustering method using the som neural network, we calibrate the student response. And we assign a grade for the student from 1 to 5. To test and evaluate the system, we used the University of North Texas Database, which lists 30 students in 12 assignments on the Computer Science Basics Test
فایل: ّFile: Download فایل